3 ms·

I would like the WHO to be as transparent as we demand China to be (and perhaps they are): At https://www.who.int/emergencies/diseases/novel-coronavirus-2019/

I would like the WHO to be as transparent as we demand China to be (and perhaps they are): At https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/early-investigations https://www.who.int/emergencies/diseases/novel-coronavirus-2... , you can see the standardized protocols for the hospitals dealing with 2019-nCov, such as the anonymizing CRF (Case Record Form): https://www.who.int/docs/default-source/coronaviruse/who-ncov-crf.pdf https://www.who.int/docs/default-source/coronaviruse/who-nco... Which describes both the CORE and DAILY forms for an electronic database. Some selected sentences from the pdf: === * The CRF is designed to collect data obtained through examination, interview and review of hospital notes. Data may be collected retrospectively if the patient is enrolled after the admission date. * DO NOT INPUT ANY PATIENT IDENTIFIERS: THIS INCLUDES NAMES, ADDRESSES, DATE OF BIRTH OR PLACE OF BIRTH. * Step 1: Contact EDCARN@who.int to become a contributor to the nCoV global platform. * Step 2: You will be contacted by ISARIC, platform manager, for assignment informational pack and instructions on how to use the REDCap nCoV platform. [...] * If your site would like to collect data independently, establishment of locally hosted database is possible. * Standard reports will be provided on regular basis to all contributors. Additional analysis for operational public health purposes will be determined by an independent WHO clinical advisory group. === As you can see, the only way to get this anonymized data, is to be a contributor, and to be a contributor, you'd have to be a hospital, or somehow be granted access by WHO. Remember the live outbreak map by John Hopkins? It also displays the number of people that recovered, the dates and definition of recovery duration are helpful in estimating mortality. Note that the other maps and graphs online only publish contaminations and deaths, which is not useful for calculating mortality. Why does John Hopkins deserve access, and why can I not get access, if I wish to fit a statistical model to the epidemic observations?

Accujack 7y ago
You can. See this page: https://www.who.int/healthinfo/statistics/en/ https://www.who.int/healthinfo/statistics/en/ Use their API or the web search, however you like. If you're going to complain about not having up to the minute data available for 2019-nCov, keep in mind that they're not getting the data submitted regularly from China, it's all in daily reports rather than from hospitals.
- DoctorOetker 7y ago
  would you mind sharing a more precise link to the actual 2019-nCoV CRF database? I want to independently recalculate mortalities
  - Accujack 7y ago
    There's not much point, because the data WHO has on the outbreak is being collated from Chinese sources who probably aren't telling the truth. WHO doesn't have a lot of people on the ground in Wuhan and the rest of China feeding them information. I'm sure there's a database for this outbreak in there somewhere, but I'm also sure the data presently in it is useless. It's likely that at this point the Chinese government doesn't even have accurate data.
    - DoctorOetker 7y ago
      >There's not much point, because the data WHO has on the outbreak is being collated from Chinese sources who probably aren't telling the truth. 1) No dataset is perfect, and everything leaves traces, so clever people can and do constantly deduce insights from apparently worthless data. 2) It is in the interest of the Chinese to cooperate with submitting these CRF's to their best ability. 3) The little data the WHO does release to the public seems quite accurate for an outbreak, if you look at the log-linear plots and so on. (I am not saying the "known infections" represent the actual number of infections, just that it looks like accurate reporting of exactly that "known infections", and if one reads the forms you see it was designed to take into account overcapacity effects, like sending a patient back home, forwarding to a different hospital, etc...). >I'm sure there's a database for this outbreak in there somewhere, but I'm also sure the data presently in it is useless. What do you mean with "in there somewhere" ? 1) If you mean the WHO has the database, then of course the WHO has the CRF database, that's what my original message pointed out. 2) If with "in there somewhere" you meant that it's publically available on their website, then no, it isn't it explicitly states it isn't: " State Parties are invited to contribute Anonymized nCoV Data to the nCoV Data Platform. State Parties should please contact WHO at EDCARN@who.int to obtain more information about, including log-in credentials for, the nCoV Platform. To preserve the security and confidentiality of the Anonymized nCoV Data, State Parties are respectfully requested to take all necessary measures to protect their respective log-in credentials and passwords to the nCoV Data Platform. " So they are sitting on the data as I originally claimed. On some of the clearly popularity boosted "contra-infodemic" threads, they even openly admit they can't give more detailed information even though they have access, when people ask them what they base their numbers on. [throughout the rest of the threads they typically go to great lengths to give the impression they work on the same aggregate numbers you and I can publically see, as if there is no censorship on the anonymized CRF data] Also, why would I care about live to the minute data? I am looking for the actual data (noisy or not), and I don't care if it's delayed by a few days. As I said, I am would like to build a statistical model. As a lesson, perhaps not for this epidemic, but then the next: if the average chinese person can afford a smartphone, then surely they can afford a couple of UV-C LED's with a battery pack, with the LED's shining through a reflective manifold (think a pipe bent back and forth with U-bends), then breathing airflow together with proper dose (X mJ per square cm) should sterilize microbes, viruses in the air. Then it just needs to be recharged, instead of trying to manufacture 61Mega masks a day. The air sterilizer (to be attached to a face mask) could be reused for new epidemics every 5 or 10 years (in which time they might have bought 3 or 4 cell phones...) EDIT: I see on change.org that there have been petitions towards the WHO on other issues before, perhaps someone should submit a petition to release the CRF database to the public domain. And it's not a question of hosting bandwidth, since the WHO can publish cryptographic hashes, and put up torrent magnets...

Accujack 7y ago

You can. See this page: https://www.who.int/healthinfo/statistics/en/ https://www.who.int/healthinfo/statistics/en/ Use their API or the web search, however you like. If you're going to complain about not having up to the minute data available for 2019-nCov, keep in mind that they're not getting the data submitted regularly from China, it's all in daily reports rather than from hospitals.

DoctorOetker 7y ago
would you mind sharing a more precise link to the actual 2019-nCoV CRF database? I want to independently recalculate mortalities
- Accujack 7y ago
  There's not much point, because the data WHO has on the outbreak is being collated from Chinese sources who probably aren't telling the truth. WHO doesn't have a lot of people on the ground in Wuhan and the rest of China feeding them information. I'm sure there's a database for this outbreak in there somewhere, but I'm also sure the data presently in it is useless. It's likely that at this point the Chinese government doesn't even have accurate data.
  - DoctorOetker 7y ago
    >There's not much point, because the data WHO has on the outbreak is being collated from Chinese sources who probably aren't telling the truth. 1) No dataset is perfect, and everything leaves traces, so clever people can and do constantly deduce insights from apparently worthless data. 2) It is in the interest of the Chinese to cooperate with submitting these CRF's to their best ability. 3) The little data the WHO does release to the public seems quite accurate for an outbreak, if you look at the log-linear plots and so on. (I am not saying the "known infections" represent the actual number of infections, just that it looks like accurate reporting of exactly that "known infections", and if one reads the forms you see it was designed to take into account overcapacity effects, like sending a patient back home, forwarding to a different hospital, etc...). >I'm sure there's a database for this outbreak in there somewhere, but I'm also sure the data presently in it is useless. What do you mean with "in there somewhere" ? 1) If you mean the WHO has the database, then of course the WHO has the CRF database, that's what my original message pointed out. 2) If with "in there somewhere" you meant that it's publically available on their website, then no, it isn't it explicitly states it isn't: " State Parties are invited to contribute Anonymized nCoV Data to the nCoV Data Platform. State Parties should please contact WHO at EDCARN@who.int to obtain more information about, including log-in credentials for, the nCoV Platform. To preserve the security and confidentiality of the Anonymized nCoV Data, State Parties are respectfully requested to take all necessary measures to protect their respective log-in credentials and passwords to the nCoV Data Platform. " So they are sitting on the data as I originally claimed. On some of the clearly popularity boosted "contra-infodemic" threads, they even openly admit they can't give more detailed information even though they have access, when people ask them what they base their numbers on. [throughout the rest of the threads they typically go to great lengths to give the impression they work on the same aggregate numbers you and I can publically see, as if there is no censorship on the anonymized CRF data] Also, why would I care about live to the minute data? I am looking for the actual data (noisy or not), and I don't care if it's delayed by a few days. As I said, I am would like to build a statistical model. As a lesson, perhaps not for this epidemic, but then the next: if the average chinese person can afford a smartphone, then surely they can afford a couple of UV-C LED's with a battery pack, with the LED's shining through a reflective manifold (think a pipe bent back and forth with U-bends), then breathing airflow together with proper dose (X mJ per square cm) should sterilize microbes, viruses in the air. Then it just needs to be recharged, instead of trying to manufacture 61Mega masks a day. The air sterilizer (to be attached to a face mask) could be reused for new epidemics every 5 or 10 years (in which time they might have bought 3 or 4 cell phones...) EDIT: I see on change.org that there have been petitions towards the WHO on other issues before, perhaps someone should submit a petition to release the CRF database to the public domain. And it's not a question of hosting bandwidth, since the WHO can publish cryptographic hashes, and put up torrent magnets...